This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
Increased incidence and size of cavum septum pellucidum in children with chromosome 22q11.2 deletion syndrome
Dec 19, 2022
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are encouraged to visit:
Author's personal copy
Increased incidence and size of cavum septum pellucidum in children with chromosome 22q11.2 deletion syndrome
Elliott A. Beatona,b, Yufeng Qinb,c, Vy Nguyena,b, Joel Johnsona,b, Joseph D. Pinterd, Tony J. Simona,b, a Department of Psychiatry, University of California, Davis, CA, USA b UC Davis M.I.N.D. Institute, University of California, Davis, Sacramento, CA, USA c Department of Child Development, Children's Hospital, Zhejiang University School of Medicine, Hangzhou, China d Department of Pediatrics, Division of Neurology, Oregon Health and Science University, Portland, OR, USA
a b s t r a c ta r t i c l e i n f o
Article history: Received 3 April 2009 Received in revised form 12 August 2009 Accepted 17 October 2009
Keywords: Genetic disorder Pediatric neuroimaging Brain volumetric methods Velocardiofacial syndrome (VCFS) DiGeorge syndrome
Chromosome 22q11.2 deletion syndrome (22q11.2DS) is a result of a hemizygotic microdeletion that results in a variety of impairments in children including greater risk for psychiatric ailments in adulthood. We used high-resolution magnetic resonance imaging to accurately quantify the length and, for the first time, volume, of the cavum septum pellucidum (CSP) in children aged 7 to 14 years with 22q11.2DS and typically developing (TD) controls. Significantly greater anteroposterior length and greater CSP volumes were found in children with 22q11.2DS compared with controls. Furthermore, the largest CSP were found only in the 22q11.2DS group and with a much higher incidence than previously reported in the literature. Given the significant midline anomalies in the brains of those affected by 22q11.2DS, large CSP may be a biomarker of atypical brain development. The implication of these larger CSP for cognitive and behavioral development is a topic in need of further investigation.
© 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
One of the earliest developmental processes in brain is the formation of the ventricular system and the associated septum separating the two chambers (Galarza et al., 2004). Typically, from months 2 to 5 of embryonic and fetal development, the lateral ventri- cles elongate and expand laterally away from the midline along with the expansion of the cerebral cortex. The anterior, posterior, and inferior horns as well as the bodies of the lateral ventricles become delineated and larger through month 7 of development. The septum pellucidum forms a medial wall between the body and the anterior horn of the lateral ventricles. Incomplete fusion of the laminae can manifest as one or two separate cavities: a cavum septum pellucidum (CSP) and a cavum vergae (CV).
Anatomically, the CSP is defined anteriorly by the genu of the corpus callosum, superiorly by the body of the corpus callosum, posteriorly by the anterior limb and pillars of the fornix, and inferiorly
by the rostrum of the corpus callosum and the anterior commissure. In prenatal month 5, anterior to posterior and superior to inferior consolidation of the corpus callosum begins. The rostrum of the corpus callosum links the genu and the lamina terminalis while the fornix remains relatively stationary and the forceps minor grow to the frontal lobes by month 7. As callosal consolidation occurs, the leaflets of the septum pellucidum are drawn together and towards the lamina terminalis, closing the CSP from rostrum to fornix. Typically, the more anterior CSP is separated from the posterior CV by the anterior columns of the fornix. If the fornix is insufficiently fused with the corpus callosum, the CSP and CV will form into one continuous cavity (Born et al., 2004). In 15% of typically developing (TD) infants, the laminae fuse within 1 month post-partum, with the majority (85%) showing laminae fusion within 6 months.
The mechanisms by which the septum pellucidum closes and by which a CSP is maintained are still not completely understood (Shashi et al., 2004), but fusion of the laminae depends on the normal development of surrounding structures, particularly the hippocampus and the corpus callosum (Sarwar, 1989). Galarza et al. (2004) suggest several possible modulators of CSP maintenance, including primary atrophy in the form of reduced frontal and temporal lobe volumes and overall hemispheric volume reduction with ventricular enlargement. Support for this mechanism includes the common clinical neuroima- ging correlation of CSP with brain anomalies characterized by global decrease in cerebral mass, such as in the pachygyria–lissencephaly spectrum and non-specific microcephaly (personal observation, JP). The proposal that laterally applied pressure, such as would be
Psychiatry Research: Neuroimaging 181 (2010) 108–113
Abbreviations: 22q11.2DS, chromosome 22q11.2 deletion syndrome; CSP, cavum septum pellucidum; TD, typically developing; IQ, intelligence quotient; CV, cavum vergae; SD, standard deviation; WISC, Wechsler Intelligence Scale for Children; FSIQ, full scale intelligence quotient; PO, Perceptual Organization; PRI, Perceptual Reasoning Index; VCI, Verbal Comprehension Index; CHOP, Children's Hospital of Philadelphia; HUP, Hospital of the University of Pennsylvania; UCD, University of California, Davis; χ2, chi-square. Corresponding author. UC Davis M.I.N.D. Institute, 2825 50th St., Sacramento, CA
95817, USA. Tel.: +1 916 703 0407; fax: +1 916 734 3384. E-mail address: [email protected] (T.J. Simon).
0925-4927/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pscychresns.2009.10.009
Contents lists available at ScienceDirect
Psychiatry Research: Neuroimaging
j ourna l homepage: www.e lsev ie r.com/ locate /psychresns
Author's personal copy
expected from increasing lobar volumes in normal brain develop- ment, can close the CSP is suggested by the observation of transient closure in a premature baby with hydrocephalus resulting from an intraventricular bleed (Needelman et al., 2006). Reduced connectivity (with resultant decreased tractional force exerted in an anterolateral direction) between the genu of the corpus callosum (the anterior border of the septum pellucidum) and the frontal lobes might also result in the maintenance of a CSP anteriorly. Since the corpus callosum also provides the superior attachment of the septum pellucidum, aberrant development leading to more lateral displace- ment could also result in less transmission of mass from superolateral cortical structures through themidline structures, which could in turn lead to maintenance of a CSP as well. CSP has also been noted with increased frequency in developmentally delayed individuals, suggest- ing it is commonly related to cerebral dysgenesis of many types (Bodensteiner and Schaefer, 1997). Indeed, it has been suggested that a wide CSP may be a non-specific marker for disturbed development (Bodensteiner and Schaefer, 1990).
We have informally observed enlarged CSP in a large number of children with chromosome 22q11.2 deletion syndrome (22q11.2DS) in the course of collecting and reviewing structural brain imaging data over the last several years. van Amelsvoort et al. (2001) report a CSP/ CV incidence rate of 40% in adults with 22q11.2DS versus a matched control group. Consistent with this, Campbell et al. (2006) reported a 69% incidence rate of midline anomalies, in particular those of CSP/CV, in children with 22q11.2DS versus 35% of sibling controls. Shashi et al. (2004) reported the presence of CSP in 4 of 13 children with 22q11.2DS making it the most common midline brain anomaly in their sample of non-psychotic children with the chromosomal deletion. 22q11.2DS encompasses the phenotypes of DiGeorge (1965), velocardiofacial (Shprintzen et al., 1978), and several other syndromes and is caused by hemizygous 1.5–3.0 Mb interstitial deletions on the q11 band of chromosome 22 (Driscoll et al., 1992). 22q11.2DS prevalence is between 1:2000 and 1:5000 live births (Oskarsdottir et al., 2004; Shprintzen, 2008) and is characterized by T-cell abnormalities, cleft palate, heart defects, facial dysmorphisms, and neonatal hypocalcemia (McDonald-McGinn et al., 1999; Antshel et al., 2005). Impairments or delays in language production and comprehension, visuospatial and numerical processing (Moss et al., 1999; Swillen et al., 1999; Bearden et al., 2001; Woodin et al., 2001; Simon et al., 2005a) and executive function (Sobin et al., 2004; Bish et al., 2005) are common, with mean IQ typically ranging from 70 to 85 (Woodin et al., 2001).
Given our recently stated position that midline anomalies of the brain may be at the root of many of the cognitive impairments experienced by those with 22q11.2DS (Simon, 2008), it may be that further understanding of this phenomenon could highlight an important early detectable biomarker for later neurocognitive dysfunction or even help to explain the mechanism by which the later impairments are created. However, to the best of our knowledge, specific quantification of CSP volumetric variability in children with 22q11.2DS has not been undertaken. Furthermore, most analytical studies of CSP have reportedmeasurements of length andwidth (Born et al., 2004; Filipovi et al., 2004), but few groups have examined CSP in volumetric terms (Crippa et al., 2006; Brisch et al., 2007). Since CSP is an anomaly that consists of a complex three-dimensional volume, strictly linear parameters such as length and width are unable to fully characterize its extent. Thus, volumetric measurements provide more accurate information than linear measurements about the true size of the cavum (Crippa et al., 2006).
The purpose of the current study was to both replicate and extend the existing literature on brain dysmorphia with a specific focus on CSP in children with 22q11.2DS. To this end, we used high-resolution magnetic resonance imaging (MRI) to specifically quantify and compare CSP volumes in children with chromosome 22q11.2DS and TD controls of a similar age. We hypothesized that CSP would be
present more often in children with 22q11.2DS versus TD children. We also hypothesized that CSP lengths and volumes would be larger in children with 22q11.2DS versus typical controls.
2. Methods
2.1. Participants
Participants were 45 children with 22q11.2DS as confirmed by fluo- rescence in-situ hybridization (20male, 25 female;mean age=10 years, 5 months, S.D.=1 year, 11 months) and 35 TD children (22 males, 13 females; mean age=10 years, 10 months, S.D.=2 years, 2 months) recruited at the Children's Hospital of Philadelphia (CHOP), the Hospital of the University of Pennsylvania (HUP), and at the University of California, Davis (UCD). After description of the study, assent was obtained from the child participants and written consent was obtained from guardians as approved by the Institutional Review Board of CHOP, HUP or UCD. The mean age difference was not significant between groups (t(78)=0.98, P=0.33) and the groups did not significantly differ in terms of gender composition (χ2(1)=2.68, P=0.10). Children recruited at UCD and HUPwere administered theWechsler Intelligence Scale for Children, version 4 (WISC 4) (Wechsler, 2003) and children recruited at CHOP were administered the WISC 3 (Wechsler, 1991). As this was a sample of convenience, the Perceptual Organization (PO) from theWISC-3 andPerceptual Reasoning Index (PRI) fromtheWISC-4 as well as the Verbal Comprehension Index (VCI) from the WISC-3 and the VCI from the WISC-4 were treated as the same measures for the purpose of assessing participants' cognitive function. IQmeasures were not available for six children with chromosome 22q11.2DS and for nine TD children.
MeanFull Scale, Verbal (VCorVCI), and Perceptual (POor PRI) scores were compared between the 22q11.2DS and TD groups with univariate analyses of variance (ANOVA) with gender and age in months as covariates. Full scale [F(1,65)=128.59, P<0.0001; 22q11.2DS mean= 76.68, S.D.=11.77; TD mean=110.04, S.D.=11.52], Verbal Compre- hension [F(1,65)=75.37, P<0.0001; 22q11.2DS mean=81.12, S.D.= 13.30; TD mean=111.32, S.D.=14.18], and Perceptual Organization/ Reasoning [F(1,65)=104.56, P<0.0001; 22q11.2DS mean=76.45, S.D.=12.55; TD mean=108.88, S.D.=11.63] scores were lower in the 22q11.2DS group than the TD children. There were no significant effects of gender or age in months on any of the IQ measures [P values>0.15].
2.2. Brain image acquisition
This was a retrospective study that aimed to take advantage of a relatively large number of brain image datasets collected at three separate institutions. Thirty-five scans (n=18 22q11.2DS, n=17 TD) were obtained at the Children's Hospital at Philadelphia, 18 scans (n=12 22q11.2DS, n=6 TD) at the Hospital of the University of Pennsylvania, and 27 scans (n=15 22q11.2DS, n=12 TD) at the University of California, Davis Medical Center. We acknowledge that there are limitations inherent with combining multi-site data sets. Optimally, we would have intraclass correlation coefficients for a sub- groupof participants or for a single phantombut given the retrospective nature of the study but this was not possible.
Three-dimensional high-resolution T1-weighted structural scans were acquired using magnetic resonance imaging (MRI) at three sep- arate institutions. All three sites utilized a magnetization prepared rapid gradient echo (MP-RAGE) sequence for image acquisition. At the Children's Hospital of Philadelphia, a 1.5-T Siemens MAGNETOM Vision scanner (Siemens Medical Solutions, Erlangen, Germany) was used with the following parameters: repetition time (TR)=1.97 s, echo time (TE)=4 s, flip angle=12°, number of excitations=1, matrix size=256×256, slice thickness=1.0 mm, 160 sagittal slices, in-plane resolution=1×1 mm. At the Hospital of the University of
109E.A. Beaton et al. / Psychiatry Research: Neuroimaging 181 (2010) 108–113
Author's personal copy
Pennsylvania, a 3.0-T Siemens MAGNETOM Vision scanner (Siemens Medical Solutions, Erlangen, Germany) was used with the following parameters: TR=1.62 s, TE=3.87 s, flip angle=15°, number of excitations=1, matrix size=192×256, slice thickness=1.0 mm, 160 sagittal slices, in-plane resolution=1×1 mm. At the University of California, Davis, a 3.0-T SiemensMAGNETOMVision scanner (Siemens Medical Solutions, Erlangen, Germany) was used with the following parameters: TR=1.82 s, TE=2.93 s, flip angle=12°, number of excitations=1, matrix size=256×256, slice thickness=1 mm, 160 sagittal slices, in-plane resolution=1×1 mm.
2.3. Image analysis
Images were transferred to a workstation for preprocessing and analysis. Image sets were aligned to the anterior commissure and posterior commissure (AC–PC) plane using Analyze 7.5 software (Biomedical Imaging Resource, Mayo Foundation, Rochester, MN). All image tracing was performed by Y.Q. and V.N. with a high degree of inter-rater reliability on measures of CSP length (r2=0.99, P<0.0001) and volume (r2=0.916, P<0.0001).
Anteroposterior CSP length can be measured by summing the number of consecutive 1-mm slices through the coronal plane where a CSP is visible (Nopoulos et al., 2000). For example, a CSP that could be seen across six coronal slices would be approximately 6 mm long. It should be noted that accurate determination of length might be limited by acquisition of partial image volumes using this method. Participants' anteroposterior CSP length was determined using this method and accordingly, participants were categorized into five groups following Nopoulos et al. (2000). Categories were defined in terms of the number of 1-mm slices (i.e. length) visible in the coronal plane. A CSP evident from 1 to 4 slices was labeled Normal; 5 to 6 slices as Borderline; 7 to 10 slices as Abnormal. If no CSP was evident, these cases were labeled as None.
Next, to gain a more accurate measure of CSP volume, the visible CSP area was traced on each 1 mm slice through the coronal plane according to predefined boundaries using Multitracer (UCLA Labora- tory of Neuroimaging, Los Angeles, CA). We operationally defined the boundaries of the CSP as follows: anteriorly by the genu of the corpus callosum, superiorly by the body of the corpus callosum, posteriorly by the anterior limb and pillars of the fornix, and inferiorly by the rostrum of the corpus callosum and the anterior commissure. When viewed in the coronal plane, the cavum is triangular with its base at the corpus callosum (Born et al., 2004). Voxels within the tracing boundaries for each slice were then summed to calculate volume in mm3.
3. Results
3.1. Prevalence of CSP
3.1.1. Prevalence by group and gender The proportion of children with CSP of any size was greater in
children with 22q11.2DS (38 out of 45; 84.4%;χ2 (1)=7.36, P=0.007) versus TD controls (20 out of 35; 57.1%). The overall proportion of having any CSP did not statistically differ (χ2 (1)=0.076, P=0.78) betweenmales (31 out of 42; 73.8%) and females (27 out of 38; 71.1%). Within each group, the proportion of males to femaleswith any CSP did not differ fromexpectation in the22q11.2DS (χ2 (1)=0.85,P=0.44) or TD (χ2 (1)=0.57, P=0.72) groups. There was also no effect of gender on severity of CSP within the 22q11.2DS (χ2 (5)=2.13, P=0.71) or TD (χ2 (3)=2.04, P=0.56) groups.
3.1.2. Prevalence by classification In the course of analyzing the images, we extended the classification
scheme by Nopoulos et al. (2000) to include an Extreme category of CSP that is twice the anteroposterior length of the upper value of the Bor-
derline classification (i.e., 12 mm or greater). These extremely large CSP (ranging from 12 to as much as 58 mm in length) occurred in 11 out of 45 of the 22q11.2DS group. Given that this was seen in 24.4% of our 22q11.2DS sample, we felt that this length range, albeit extreme, should be considered as a new category of CSP rather than as a set of outliers whose relevance is not considered further. In fact, the percentage of children with 22q11.2DS in the Extreme category was second only to those in the Normal category. Example images of each classification category are shown in Fig. 1.
The proportion and frequencies of individuals that fell into theNone, Normal, Borderline, Abnormal, and Extreme categories are shown in Table 1. There was a significant relationship between group and overall CSP classification (χ2 (4)=13.00, P=0.009). There was no detectable CSP in 15.6% (7/45) of the 22q11.2DS group compared to 42.9% (15/35) of the TD group, and it was statisticallymore likely that a TD participant would fall into the None CSP classification (χ2 (1)=4.90, P=0.027). We detected the Normal variant CSP in 44.4% (20/45) of the 22q11.2DS children and in 37.1% (13/35) of the TD group and the groups did not differ in the likelihood of this classification (χ2 (1)=0.025, P=0.87). Borderline CSP was present in 4.4% (2/45) of the 22q11.2DS and 11.4% (4/35) of the TD group, and this proportion was also not significantly different (χ2 (1)=1.38, P=0.24) between groups. Children with 22q11.2DS were no more likely to possess an Abnormal CSP (χ2 (1)= 0.71, P=0.40) thanTD childrenwith 11.1% (5/45) of the 22q11.2DS and 8.6% (3/35) of the TD group meeting the Abnormal threshold. None of the TD children met the Extreme CSP criteria whereas 24.4% (11/45) of the children with 22q11.2DS had CSP 12 mm or longer indicating a statistically significantdifference in the likelihoodof CSP classification in the Extreme category (χ2 (1)=7.42, P=0.06).
3.2. CSP morphometry
3.2.1. CSP lengths The anteroposterior CSP lengths (i.e. sum of 1-mm slices where CSP
was evident) were compared between TD children (M=2.31; S.D.= 2.60) and those with 22q11.2DS (M=12.58; S.D.=18.64) controlling for gender, age, and scanner site using analysis of variance (ANOVA) revealing a significant main effect of Group [F(1)=10.11, P=0.002]. There were no significant effects of age [F(1)=0.58, P=0.45], gender [F(1)=0.011, P=0.92], or scanner site [F(1)=0.12, P=0.73]. Box plots of mean log-transformed CSP length by group are illustrated in Fig. 2A.
3.2.2. CSP volumes Mean CSP volumes were calculated for children with 22q11.2DS
(N=45; M=916.13 mm3; S.D.=2413.68) and TD (N=35; M= 12.83 mm3; S.D.=19.21) children. Since CSP volumes ranged from 0 to 13,054 mm3 and the distribution of volumes was extremely negatively skewed [skewness=4.89, SE=0.27], parametric analyses were not possible. Thus, we log-transformed CSP values. Cava were larger in the 22q11.2DS group than in the TD group. Univariate ANOVA comparing logCSP volumes across Groups (22q11.2DS, TD) while controlling for gender, age in months, and scanner site revealed a significant main effect of Group [F(1, 75)=14.93, P<0.001]. There were no significant effects of age [F(1,75)=1.30, P=0.26], gender [F(1, 75)=0.10, P=0.75], or scanner site [F(1, 75)=1.17, P=0.28] in relation to CSP volume. Univariate ANOVA comparing log-transformed volumes for all participants across scanner sites did not indicate a statistically significant difference [F(2,79)=1.16, P=0.32]. Box plots of mean log- transformed CSP volume…
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are encouraged to visit:
Author's personal copy
Increased incidence and size of cavum septum pellucidum in children with chromosome 22q11.2 deletion syndrome
Elliott A. Beatona,b, Yufeng Qinb,c, Vy Nguyena,b, Joel Johnsona,b, Joseph D. Pinterd, Tony J. Simona,b, a Department of Psychiatry, University of California, Davis, CA, USA b UC Davis M.I.N.D. Institute, University of California, Davis, Sacramento, CA, USA c Department of Child Development, Children's Hospital, Zhejiang University School of Medicine, Hangzhou, China d Department of Pediatrics, Division of Neurology, Oregon Health and Science University, Portland, OR, USA
a b s t r a c ta r t i c l e i n f o
Article history: Received 3 April 2009 Received in revised form 12 August 2009 Accepted 17 October 2009
Keywords: Genetic disorder Pediatric neuroimaging Brain volumetric methods Velocardiofacial syndrome (VCFS) DiGeorge syndrome
Chromosome 22q11.2 deletion syndrome (22q11.2DS) is a result of a hemizygotic microdeletion that results in a variety of impairments in children including greater risk for psychiatric ailments in adulthood. We used high-resolution magnetic resonance imaging to accurately quantify the length and, for the first time, volume, of the cavum septum pellucidum (CSP) in children aged 7 to 14 years with 22q11.2DS and typically developing (TD) controls. Significantly greater anteroposterior length and greater CSP volumes were found in children with 22q11.2DS compared with controls. Furthermore, the largest CSP were found only in the 22q11.2DS group and with a much higher incidence than previously reported in the literature. Given the significant midline anomalies in the brains of those affected by 22q11.2DS, large CSP may be a biomarker of atypical brain development. The implication of these larger CSP for cognitive and behavioral development is a topic in need of further investigation.
© 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
One of the earliest developmental processes in brain is the formation of the ventricular system and the associated septum separating the two chambers (Galarza et al., 2004). Typically, from months 2 to 5 of embryonic and fetal development, the lateral ventri- cles elongate and expand laterally away from the midline along with the expansion of the cerebral cortex. The anterior, posterior, and inferior horns as well as the bodies of the lateral ventricles become delineated and larger through month 7 of development. The septum pellucidum forms a medial wall between the body and the anterior horn of the lateral ventricles. Incomplete fusion of the laminae can manifest as one or two separate cavities: a cavum septum pellucidum (CSP) and a cavum vergae (CV).
Anatomically, the CSP is defined anteriorly by the genu of the corpus callosum, superiorly by the body of the corpus callosum, posteriorly by the anterior limb and pillars of the fornix, and inferiorly
by the rostrum of the corpus callosum and the anterior commissure. In prenatal month 5, anterior to posterior and superior to inferior consolidation of the corpus callosum begins. The rostrum of the corpus callosum links the genu and the lamina terminalis while the fornix remains relatively stationary and the forceps minor grow to the frontal lobes by month 7. As callosal consolidation occurs, the leaflets of the septum pellucidum are drawn together and towards the lamina terminalis, closing the CSP from rostrum to fornix. Typically, the more anterior CSP is separated from the posterior CV by the anterior columns of the fornix. If the fornix is insufficiently fused with the corpus callosum, the CSP and CV will form into one continuous cavity (Born et al., 2004). In 15% of typically developing (TD) infants, the laminae fuse within 1 month post-partum, with the majority (85%) showing laminae fusion within 6 months.
The mechanisms by which the septum pellucidum closes and by which a CSP is maintained are still not completely understood (Shashi et al., 2004), but fusion of the laminae depends on the normal development of surrounding structures, particularly the hippocampus and the corpus callosum (Sarwar, 1989). Galarza et al. (2004) suggest several possible modulators of CSP maintenance, including primary atrophy in the form of reduced frontal and temporal lobe volumes and overall hemispheric volume reduction with ventricular enlargement. Support for this mechanism includes the common clinical neuroima- ging correlation of CSP with brain anomalies characterized by global decrease in cerebral mass, such as in the pachygyria–lissencephaly spectrum and non-specific microcephaly (personal observation, JP). The proposal that laterally applied pressure, such as would be
Psychiatry Research: Neuroimaging 181 (2010) 108–113
Abbreviations: 22q11.2DS, chromosome 22q11.2 deletion syndrome; CSP, cavum septum pellucidum; TD, typically developing; IQ, intelligence quotient; CV, cavum vergae; SD, standard deviation; WISC, Wechsler Intelligence Scale for Children; FSIQ, full scale intelligence quotient; PO, Perceptual Organization; PRI, Perceptual Reasoning Index; VCI, Verbal Comprehension Index; CHOP, Children's Hospital of Philadelphia; HUP, Hospital of the University of Pennsylvania; UCD, University of California, Davis; χ2, chi-square. Corresponding author. UC Davis M.I.N.D. Institute, 2825 50th St., Sacramento, CA
95817, USA. Tel.: +1 916 703 0407; fax: +1 916 734 3384. E-mail address: [email protected] (T.J. Simon).
0925-4927/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pscychresns.2009.10.009
Contents lists available at ScienceDirect
Psychiatry Research: Neuroimaging
j ourna l homepage: www.e lsev ie r.com/ locate /psychresns
Author's personal copy
expected from increasing lobar volumes in normal brain develop- ment, can close the CSP is suggested by the observation of transient closure in a premature baby with hydrocephalus resulting from an intraventricular bleed (Needelman et al., 2006). Reduced connectivity (with resultant decreased tractional force exerted in an anterolateral direction) between the genu of the corpus callosum (the anterior border of the septum pellucidum) and the frontal lobes might also result in the maintenance of a CSP anteriorly. Since the corpus callosum also provides the superior attachment of the septum pellucidum, aberrant development leading to more lateral displace- ment could also result in less transmission of mass from superolateral cortical structures through themidline structures, which could in turn lead to maintenance of a CSP as well. CSP has also been noted with increased frequency in developmentally delayed individuals, suggest- ing it is commonly related to cerebral dysgenesis of many types (Bodensteiner and Schaefer, 1997). Indeed, it has been suggested that a wide CSP may be a non-specific marker for disturbed development (Bodensteiner and Schaefer, 1990).
We have informally observed enlarged CSP in a large number of children with chromosome 22q11.2 deletion syndrome (22q11.2DS) in the course of collecting and reviewing structural brain imaging data over the last several years. van Amelsvoort et al. (2001) report a CSP/ CV incidence rate of 40% in adults with 22q11.2DS versus a matched control group. Consistent with this, Campbell et al. (2006) reported a 69% incidence rate of midline anomalies, in particular those of CSP/CV, in children with 22q11.2DS versus 35% of sibling controls. Shashi et al. (2004) reported the presence of CSP in 4 of 13 children with 22q11.2DS making it the most common midline brain anomaly in their sample of non-psychotic children with the chromosomal deletion. 22q11.2DS encompasses the phenotypes of DiGeorge (1965), velocardiofacial (Shprintzen et al., 1978), and several other syndromes and is caused by hemizygous 1.5–3.0 Mb interstitial deletions on the q11 band of chromosome 22 (Driscoll et al., 1992). 22q11.2DS prevalence is between 1:2000 and 1:5000 live births (Oskarsdottir et al., 2004; Shprintzen, 2008) and is characterized by T-cell abnormalities, cleft palate, heart defects, facial dysmorphisms, and neonatal hypocalcemia (McDonald-McGinn et al., 1999; Antshel et al., 2005). Impairments or delays in language production and comprehension, visuospatial and numerical processing (Moss et al., 1999; Swillen et al., 1999; Bearden et al., 2001; Woodin et al., 2001; Simon et al., 2005a) and executive function (Sobin et al., 2004; Bish et al., 2005) are common, with mean IQ typically ranging from 70 to 85 (Woodin et al., 2001).
Given our recently stated position that midline anomalies of the brain may be at the root of many of the cognitive impairments experienced by those with 22q11.2DS (Simon, 2008), it may be that further understanding of this phenomenon could highlight an important early detectable biomarker for later neurocognitive dysfunction or even help to explain the mechanism by which the later impairments are created. However, to the best of our knowledge, specific quantification of CSP volumetric variability in children with 22q11.2DS has not been undertaken. Furthermore, most analytical studies of CSP have reportedmeasurements of length andwidth (Born et al., 2004; Filipovi et al., 2004), but few groups have examined CSP in volumetric terms (Crippa et al., 2006; Brisch et al., 2007). Since CSP is an anomaly that consists of a complex three-dimensional volume, strictly linear parameters such as length and width are unable to fully characterize its extent. Thus, volumetric measurements provide more accurate information than linear measurements about the true size of the cavum (Crippa et al., 2006).
The purpose of the current study was to both replicate and extend the existing literature on brain dysmorphia with a specific focus on CSP in children with 22q11.2DS. To this end, we used high-resolution magnetic resonance imaging (MRI) to specifically quantify and compare CSP volumes in children with chromosome 22q11.2DS and TD controls of a similar age. We hypothesized that CSP would be
present more often in children with 22q11.2DS versus TD children. We also hypothesized that CSP lengths and volumes would be larger in children with 22q11.2DS versus typical controls.
2. Methods
2.1. Participants
Participants were 45 children with 22q11.2DS as confirmed by fluo- rescence in-situ hybridization (20male, 25 female;mean age=10 years, 5 months, S.D.=1 year, 11 months) and 35 TD children (22 males, 13 females; mean age=10 years, 10 months, S.D.=2 years, 2 months) recruited at the Children's Hospital of Philadelphia (CHOP), the Hospital of the University of Pennsylvania (HUP), and at the University of California, Davis (UCD). After description of the study, assent was obtained from the child participants and written consent was obtained from guardians as approved by the Institutional Review Board of CHOP, HUP or UCD. The mean age difference was not significant between groups (t(78)=0.98, P=0.33) and the groups did not significantly differ in terms of gender composition (χ2(1)=2.68, P=0.10). Children recruited at UCD and HUPwere administered theWechsler Intelligence Scale for Children, version 4 (WISC 4) (Wechsler, 2003) and children recruited at CHOP were administered the WISC 3 (Wechsler, 1991). As this was a sample of convenience, the Perceptual Organization (PO) from theWISC-3 andPerceptual Reasoning Index (PRI) fromtheWISC-4 as well as the Verbal Comprehension Index (VCI) from the WISC-3 and the VCI from the WISC-4 were treated as the same measures for the purpose of assessing participants' cognitive function. IQmeasures were not available for six children with chromosome 22q11.2DS and for nine TD children.
MeanFull Scale, Verbal (VCorVCI), and Perceptual (POor PRI) scores were compared between the 22q11.2DS and TD groups with univariate analyses of variance (ANOVA) with gender and age in months as covariates. Full scale [F(1,65)=128.59, P<0.0001; 22q11.2DS mean= 76.68, S.D.=11.77; TD mean=110.04, S.D.=11.52], Verbal Compre- hension [F(1,65)=75.37, P<0.0001; 22q11.2DS mean=81.12, S.D.= 13.30; TD mean=111.32, S.D.=14.18], and Perceptual Organization/ Reasoning [F(1,65)=104.56, P<0.0001; 22q11.2DS mean=76.45, S.D.=12.55; TD mean=108.88, S.D.=11.63] scores were lower in the 22q11.2DS group than the TD children. There were no significant effects of gender or age in months on any of the IQ measures [P values>0.15].
2.2. Brain image acquisition
This was a retrospective study that aimed to take advantage of a relatively large number of brain image datasets collected at three separate institutions. Thirty-five scans (n=18 22q11.2DS, n=17 TD) were obtained at the Children's Hospital at Philadelphia, 18 scans (n=12 22q11.2DS, n=6 TD) at the Hospital of the University of Pennsylvania, and 27 scans (n=15 22q11.2DS, n=12 TD) at the University of California, Davis Medical Center. We acknowledge that there are limitations inherent with combining multi-site data sets. Optimally, we would have intraclass correlation coefficients for a sub- groupof participants or for a single phantombut given the retrospective nature of the study but this was not possible.
Three-dimensional high-resolution T1-weighted structural scans were acquired using magnetic resonance imaging (MRI) at three sep- arate institutions. All three sites utilized a magnetization prepared rapid gradient echo (MP-RAGE) sequence for image acquisition. At the Children's Hospital of Philadelphia, a 1.5-T Siemens MAGNETOM Vision scanner (Siemens Medical Solutions, Erlangen, Germany) was used with the following parameters: repetition time (TR)=1.97 s, echo time (TE)=4 s, flip angle=12°, number of excitations=1, matrix size=256×256, slice thickness=1.0 mm, 160 sagittal slices, in-plane resolution=1×1 mm. At the Hospital of the University of
109E.A. Beaton et al. / Psychiatry Research: Neuroimaging 181 (2010) 108–113
Author's personal copy
Pennsylvania, a 3.0-T Siemens MAGNETOM Vision scanner (Siemens Medical Solutions, Erlangen, Germany) was used with the following parameters: TR=1.62 s, TE=3.87 s, flip angle=15°, number of excitations=1, matrix size=192×256, slice thickness=1.0 mm, 160 sagittal slices, in-plane resolution=1×1 mm. At the University of California, Davis, a 3.0-T SiemensMAGNETOMVision scanner (Siemens Medical Solutions, Erlangen, Germany) was used with the following parameters: TR=1.82 s, TE=2.93 s, flip angle=12°, number of excitations=1, matrix size=256×256, slice thickness=1 mm, 160 sagittal slices, in-plane resolution=1×1 mm.
2.3. Image analysis
Images were transferred to a workstation for preprocessing and analysis. Image sets were aligned to the anterior commissure and posterior commissure (AC–PC) plane using Analyze 7.5 software (Biomedical Imaging Resource, Mayo Foundation, Rochester, MN). All image tracing was performed by Y.Q. and V.N. with a high degree of inter-rater reliability on measures of CSP length (r2=0.99, P<0.0001) and volume (r2=0.916, P<0.0001).
Anteroposterior CSP length can be measured by summing the number of consecutive 1-mm slices through the coronal plane where a CSP is visible (Nopoulos et al., 2000). For example, a CSP that could be seen across six coronal slices would be approximately 6 mm long. It should be noted that accurate determination of length might be limited by acquisition of partial image volumes using this method. Participants' anteroposterior CSP length was determined using this method and accordingly, participants were categorized into five groups following Nopoulos et al. (2000). Categories were defined in terms of the number of 1-mm slices (i.e. length) visible in the coronal plane. A CSP evident from 1 to 4 slices was labeled Normal; 5 to 6 slices as Borderline; 7 to 10 slices as Abnormal. If no CSP was evident, these cases were labeled as None.
Next, to gain a more accurate measure of CSP volume, the visible CSP area was traced on each 1 mm slice through the coronal plane according to predefined boundaries using Multitracer (UCLA Labora- tory of Neuroimaging, Los Angeles, CA). We operationally defined the boundaries of the CSP as follows: anteriorly by the genu of the corpus callosum, superiorly by the body of the corpus callosum, posteriorly by the anterior limb and pillars of the fornix, and inferiorly by the rostrum of the corpus callosum and the anterior commissure. When viewed in the coronal plane, the cavum is triangular with its base at the corpus callosum (Born et al., 2004). Voxels within the tracing boundaries for each slice were then summed to calculate volume in mm3.
3. Results
3.1. Prevalence of CSP
3.1.1. Prevalence by group and gender The proportion of children with CSP of any size was greater in
children with 22q11.2DS (38 out of 45; 84.4%;χ2 (1)=7.36, P=0.007) versus TD controls (20 out of 35; 57.1%). The overall proportion of having any CSP did not statistically differ (χ2 (1)=0.076, P=0.78) betweenmales (31 out of 42; 73.8%) and females (27 out of 38; 71.1%). Within each group, the proportion of males to femaleswith any CSP did not differ fromexpectation in the22q11.2DS (χ2 (1)=0.85,P=0.44) or TD (χ2 (1)=0.57, P=0.72) groups. There was also no effect of gender on severity of CSP within the 22q11.2DS (χ2 (5)=2.13, P=0.71) or TD (χ2 (3)=2.04, P=0.56) groups.
3.1.2. Prevalence by classification In the course of analyzing the images, we extended the classification
scheme by Nopoulos et al. (2000) to include an Extreme category of CSP that is twice the anteroposterior length of the upper value of the Bor-
derline classification (i.e., 12 mm or greater). These extremely large CSP (ranging from 12 to as much as 58 mm in length) occurred in 11 out of 45 of the 22q11.2DS group. Given that this was seen in 24.4% of our 22q11.2DS sample, we felt that this length range, albeit extreme, should be considered as a new category of CSP rather than as a set of outliers whose relevance is not considered further. In fact, the percentage of children with 22q11.2DS in the Extreme category was second only to those in the Normal category. Example images of each classification category are shown in Fig. 1.
The proportion and frequencies of individuals that fell into theNone, Normal, Borderline, Abnormal, and Extreme categories are shown in Table 1. There was a significant relationship between group and overall CSP classification (χ2 (4)=13.00, P=0.009). There was no detectable CSP in 15.6% (7/45) of the 22q11.2DS group compared to 42.9% (15/35) of the TD group, and it was statisticallymore likely that a TD participant would fall into the None CSP classification (χ2 (1)=4.90, P=0.027). We detected the Normal variant CSP in 44.4% (20/45) of the 22q11.2DS children and in 37.1% (13/35) of the TD group and the groups did not differ in the likelihood of this classification (χ2 (1)=0.025, P=0.87). Borderline CSP was present in 4.4% (2/45) of the 22q11.2DS and 11.4% (4/35) of the TD group, and this proportion was also not significantly different (χ2 (1)=1.38, P=0.24) between groups. Children with 22q11.2DS were no more likely to possess an Abnormal CSP (χ2 (1)= 0.71, P=0.40) thanTD childrenwith 11.1% (5/45) of the 22q11.2DS and 8.6% (3/35) of the TD group meeting the Abnormal threshold. None of the TD children met the Extreme CSP criteria whereas 24.4% (11/45) of the children with 22q11.2DS had CSP 12 mm or longer indicating a statistically significantdifference in the likelihoodof CSP classification in the Extreme category (χ2 (1)=7.42, P=0.06).
3.2. CSP morphometry
3.2.1. CSP lengths The anteroposterior CSP lengths (i.e. sum of 1-mm slices where CSP
was evident) were compared between TD children (M=2.31; S.D.= 2.60) and those with 22q11.2DS (M=12.58; S.D.=18.64) controlling for gender, age, and scanner site using analysis of variance (ANOVA) revealing a significant main effect of Group [F(1)=10.11, P=0.002]. There were no significant effects of age [F(1)=0.58, P=0.45], gender [F(1)=0.011, P=0.92], or scanner site [F(1)=0.12, P=0.73]. Box plots of mean log-transformed CSP length by group are illustrated in Fig. 2A.
3.2.2. CSP volumes Mean CSP volumes were calculated for children with 22q11.2DS
(N=45; M=916.13 mm3; S.D.=2413.68) and TD (N=35; M= 12.83 mm3; S.D.=19.21) children. Since CSP volumes ranged from 0 to 13,054 mm3 and the distribution of volumes was extremely negatively skewed [skewness=4.89, SE=0.27], parametric analyses were not possible. Thus, we log-transformed CSP values. Cava were larger in the 22q11.2DS group than in the TD group. Univariate ANOVA comparing logCSP volumes across Groups (22q11.2DS, TD) while controlling for gender, age in months, and scanner site revealed a significant main effect of Group [F(1, 75)=14.93, P<0.001]. There were no significant effects of age [F(1,75)=1.30, P=0.26], gender [F(1, 75)=0.10, P=0.75], or scanner site [F(1, 75)=1.17, P=0.28] in relation to CSP volume. Univariate ANOVA comparing log-transformed volumes for all participants across scanner sites did not indicate a statistically significant difference [F(2,79)=1.16, P=0.32]. Box plots of mean log- transformed CSP volume…
Related Documents
![Prevalence of Cavum Septum Pellucidum in Alcohol Dependent ... fileSyndrome (ADS)in the year 2012 and 2013 as per ICD-10, DCR [14] and who had been admitted in the S.S. Raju Centre](https://static.cupdf.com/doc/110x72/5cb304b488c9934c708c245c/prevalence-of-cavum-septum-pellucidum-in-alcohol-dependent-adsin-the-year.jpg)
